retro

Analyzes coding sessions to propose improvements to agent environment configuration.

2|Updated Jun 21, 2026
One-click install
npx skills add https://github.com/IsKenKenYa/skills --skill retro-iskenkenya
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: retro
Source: https://github.com/IsKenKenYa/skills/tree/main/skills/in-progress/retro
Command: npx skills add https://github.com/IsKenKenYa/skills --skill retro-iskenkenya

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After a coding session, it is hard to know why the AI agent struggled—slow file navigation, missing automated checks, or bloated steering files. This Skill runs a structured retrospective on a session and turns those friction points into concrete improvements for the agent's environment. ## Core Features & Use Cases - Session Log Analysis: Reads primary sources and session logs to reconstruct what actually happened during a run. - Seven Improvement Categories: Evaluates navigation, automated checks, coding standards, global AGENTS.md, tool economy, no-op instructions, and information access. - Severity-Ranked Findings: Presents candidate improvements ordered by severity so you can act on the highest-impact items first. - Use Case: After an agent repeatedly failed lint checks that CI never ran, invoke a retrospective to discover the repo has no guardrail wired up and get a recommendation to add a pre-commit hook or CI job. ## Quick Start Ask the agent to run a retrospective on the current coding session and suggest environment improvements.

Frequently Asked Questions about retro

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a retrospective on an AI coding session?

Invoke the retro skill and specify the session to analyze; if none is given, it defaults to the current session. It reads session logs, identifies improvement candidates across seven categories, and presents them ranked by severity.

What kinds of improvements does a coding session retrospective find?

It finds navigation pointers for hard-to-locate files, missing automated checks like linting or CI guardrails, coding standard rules for the reviewer agent, oversized AGENTS.md instructions, token-inefficient tool calls, no-op steering instructions, and information access gaps.

Should coding standards be enforced by the implementation or review agent?

The review agent should enforce coding standards because it has the least context pressure—it receives a diff without needing exploration or debugging. The implementation agent carries the most context pressure and should stay focused on writing code.

When should a coding rule become an automated check instead of documentation?

Mechanical violations like banned APIs, import shapes, or file-location rules should become deterministic checks such as linter rules, pre-commit hooks, or CI jobs. Reserve CODING_STANDARDS.md for genuine judgement calls that no automated guardrail could enforce.

Can this skill be triggered automatically by the AI model?

No, it sets disable-model-invocation to true and its policy disallows implicit invocation, so it only runs when the user explicitly requests a retrospective.